Attention: Create Field Configuration Item

Add a field to an Attention field configuration, with its prompt and answer type.

Tool ID: attention_create_field_configuration_item

Run this action

Use the TypeScript SDK for one call. Use ctx.tools.execute(...) inside a Play for durable calls, scheduled calls, or CSV processing.

import { Deepline } from 'deepline';

const deepline = await Deepline.connect();
const result = await deepline.tools.execute(
  'attention_create_field_configuration_item',
  {
    "id": "id_123",
    "title": "VP of Engineering",
    "category": "CRM",
    "contents": "example",
    "promptType": "example",
    "contentType": "example"
  },
);

console.log(result.toolResponse.raw);

CLI

deepline tools execute attention_create_field_configuration_item --input '{
  "id": "id_123",
  "title": "VP of Engineering",
  "category": "CRM",
  "contents": "example",
  "promptType": "example",
  "contentType": "example"
}' --json

Example response

The SDK returns this shape at result.toolResponse.raw. The example values illustrate the response.

{
  "data": {
    "data": {
      "item": {
        "uuid": "uuid_123",
        "fieldConfigurationUUID": "field_configuration_123",
        "title": "VP of Engineering",
        "category": "CRM",
        "contents": "example",
        "promptType": "example",
        "contentType": "example",
        "allowLongResponse": true,
        "extractQuotes": true,
        "availableToAllClients": true,
        "createdAt": "2026-01-15T12:00:00Z",
        "lastModifiedAt": "2026-01-15T12:00:00Z"
      },
      "assignments": [
        {
          "uuid": "uuid_123",
          "intelligenceItemUUID": "intelligence_item_123",
          "teamUUID": "team_123",
          "category": "example",
          "createdAt": "2026-01-15T12:00:00Z",
          "lastModifiedAt": "2026-01-15T12:00:00Z"
        }
      ]
    }
  }
}

Use deepline describe attention_create_field_configuration_item --json for the latest machine-readable contract.

Input reference

Request payload for creating a single new item inside a field configuration. The REST endpoint wraps this singular request into the underlying bulk creation path (ADR-9). Creating an item fans out into one IntelligenceItemAssignment per team the parent configuration is assigned to - the response returns the list of created assignments.

Name	Type	Required	Default	Details
`payload.id`	`string`	Yes	—	UUID of the field configuration.
`payload.title`	`string`	Yes	—	Short human-readable title of the item. Trimmed. 1-128 characters.
`payload.category`	`"CRM" \| "Indicators" \| "BuiltInInsight" \| "CustomInsight"`	Yes	—	Category of the item. Allowed: `CRM`, `Indicators`, `BuiltInInsight`, `CustomInsight`.
`payload.contents`	`string`	Yes	—	Prompt text that Attention will evaluate against each conversation.
`payload.description`	`string`	No	—	Optional longer description. 0-2000 characters.
`payload.promptType`	`string`	Yes	—	Prompt family used to analyze the conversation.
`payload.contentType`	`string`	Yes	—	Expected shape of the extracted content.
`payload.responseType`	`string`	No	—	Optional response data type hint.
`payload.options`	`array`	No	—	Optional allowed response values for multiple-choice items.
`payload.dependsOn`	`array`	No	—	UUIDs of other items this item depends on.
`payload.maxValues`	`integer`	No	—	Maximum number of values allowed for multi-select items.
`payload.allowLongResponse`	`boolean`	No	—	Whether the analyzer may produce a longer free-text response.
`payload.extractQuotes`	`boolean`	No	—	Whether the analyzer should attach supporting quotes to the response.
`payload.analysisLevel`	`"conversation" \| "deal" \| "lead" \| "contact" \| "account"`	No	—	Level at which the item is evaluated. Allowed: `conversation`, `deal`, `lead`, `contact`, `account`.
`payload.metadata`	`object`	No	—	Numeric scoring metadata for a field-configuration item. Required when the item represents a numeric score - `min_score` must be less than or equal to `max_score`.
`payload.displayOrder`	`integer`	No	—	Optional starting position in the team’s ordered item list. existing assignments at or after this position are shifted by one.
`payload.requestStage`	`integer`	No	—	Optional analysis stage in which the resulting assignments are evaluated.

Output reference

Standard tool result payload.

Name	Type	Required	Default	Details
`result.data`	`object`	Yes	—	Envelope returned when a field-configuration item is created. The `data.assignments` array documents the fan-out (one entry per team the parent configuration is assigned to).
`result.data.data`	`object`	No	—	Create-item payload: the newly created item plus the fan-out `IntelligenceItemAssignment` rows - one per team the parent field configuration is assigned to.
`result.data.data.item`	`object`	No	—	A single field within a Field Configuration (internally `IntelligenceItem`). Items describe a prompt that Attention evaluates against a conversation and produces a value for. A field-configuration item is cloned as one `IntelligenceItemAssignment` per team the parent configuration is assigned to - see `IntelligenceItemAssignment`.
`result.data.data.item.uuid`	`string`	Yes	—	Unique identifier of the item.
`result.data.data.item.fieldConfigurationUUID`	`string`	Yes	—	UUID of the parent field configuration.
`result.data.data.item.title`	`string`	Yes	—	Short human-readable title of the item. Trimmed. 1-128 characters.
`result.data.data.item.category`	`"CRM" \| "Indicators" \| "BuiltInInsight" \| "CustomInsight"`	Yes	—	Category of the item. One of the intelligence-category enum values. Allowed: `CRM`, `Indicators`, `BuiltInInsight`, `CustomInsight`.
`result.data.data.item.contents`	`string`	Yes	—	Prompt text that Attention will evaluate against each conversation.
`result.data.data.item.description`	`string`	No	—	Optional longer description shown to end users. 0-2000 characters.
`result.data.data.item.promptType`	`string`	Yes	—	Prompt family used to analyze the conversation.
`result.data.data.item.contentType`	`string`	Yes	—	Expected shape of the extracted content.
`result.data.data.item.responseType`	`string`	No	—	Optional response data type hint.
`result.data.data.item.options`	`array`	No	—	Optional allowed response values when the item is a multiple-choice field.
`result.data.data.item.dependsOn`	`array`	No	—	UUIDs of other items this item depends on.
`result.data.data.item.maxValues`	`integer`	No	—	Maximum number of values allowed for multi-select items.
`result.data.data.item.allowLongResponse`	`boolean`	Yes	—	Whether the analyzer may produce a longer free-text response.
`result.data.data.item.extractQuotes`	`boolean`	Yes	—	Whether the analyzer should attach supporting quotes to the response.
`result.data.data.item.availableToAllClients`	`boolean`	Yes	—	Internal flag marking the item as globally available.
`result.data.data.item.analysisLevel`	`"conversation" \| "deal" \| "lead" \| "contact" \| "account"`	No	—	Level at which the item is evaluated. Allowed: `conversation`, `deal`, `lead`, `contact`, `account`.
`result.data.data.item.metadata`	`object`	No	—	Numeric scoring metadata for a field-configuration item. Required when the item represents a numeric score - `min_score` must be less than or equal to `max_score`.
`result.data.data.item.metadata.min_score`	`number`	No	—	Minimum allowed score value. Format: `double`.
`result.data.data.item.metadata.max_score`	`number`	No	—	Maximum allowed score value. Format: `double`.
`result.data.data.item.metadata.middle_score`	`number \| null`	No	—	Optional midpoint score value, used by scoring heuristics. Format: `double`.
`result.data.data.item.metadata.min_criteria`	`string`	No	—	Human-readable description of what the minimum score means.
`result.data.data.item.metadata.max_criteria`	`string`	No	—	Human-readable description of what the maximum score means.
`result.data.data.item.metadata.middle_criteria`	`string \| null`	No	—	Human-readable description of what the middle score means.
`result.data.data.item.createdAt`	`string`	Yes	—	Timestamp when the item was created. Format: `date-time`.
`result.data.data.item.lastModifiedAt`	`string`	Yes	—	Timestamp when the item was last updated. Format: `date-time`.
`result.data.data.assignments`	`array`	No	—	Assignment rows created as part of the fan-out. For a configuration assigned to N teams, N assignments are returned.
`result.data.data.assignments[].uuid`	`string`	Yes	—	Unique identifier of the assignment row.
`result.data.data.assignments[].intelligenceItemUUID`	`string`	Yes	—	UUID of the underlying field-configuration item this assignment points at.
`result.data.data.assignments[].teamUUID`	`string`	Yes	—	UUID of the team this assignment applies to.
`result.data.data.assignments[].organizationUUID`	`string`	No	—	UUID of the organization this assignment belongs to.
`result.data.data.assignments[].category`	`string`	Yes	—	Category of the underlying item (copied from the item at assignment time).
`result.data.data.assignments[].displayOrder`	`integer`	No	—	Position of this assignment in the team’s ordered list of items.
`result.data.data.assignments[].requestStage`	`integer`	No	—	Analysis stage in which this assignment is evaluated.
`result.data.data.assignments[].createdAt`	`string`	Yes	—	Timestamp when the assignment was created. Format: `date-time`.
`result.data.data.assignments[].lastModifiedAt`	`string`	Yes	—	Timestamp when the assignment was last updated. Format: `date-time`.
`result.meta`	`object`	No	—	Additional response metadata (status, paging).

Deepline cost

Pricing model: fixed (per call).
Estimated Deepline credits: 0 per pricing unit.
Provider-native pricing may still exist outside Deepline credit billing.

​Run this action

​CLI

​Example response

​Input reference

​Output reference

​Deepline cost

​Related documentation

Run this action

CLI

Example response

Input reference

Output reference

Deepline cost

Related documentation